# Generalized Few-Shot Video Classification with Video Retrieval and Feature Generation

arXiv: https://arxiv.org/abs/2007.04755 · HTML: https://mlsanity.com/paper/2007.04755/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Jul 2020.

## Built on by (most cited first, 6 of 6)

- [Compound Prototype Matching for Few-shot Action Recognition](https://mlsanity.com/paper/2207.05515/) (2022)
- [HyRSM++: Hybrid Relation Guided Temporal Set Matching for Few-shot Action Recognition](https://mlsanity.com/paper/2301.03330/) (2023)
- [A Comprehensive Review of Few-shot Action Recognition](https://mlsanity.com/paper/2407.14744/) (2024)
- [Text-to-feature diffusion for audio-visual few-shot learning](https://mlsanity.com/paper/2309.03869/) (2023)
- [Rethinking matching-based few-shot action recognition](https://mlsanity.com/paper/2303.16084/) (2023)
- [All Eyes on the Workflow: Automated and Efficient Event Discovery from Video Streams](https://mlsanity.com/paper/2604.22476/) (2026)

## Abstract (arXiv)

Few-shot learning aims to recognize novel classes from a few examples. Although significant progress has been made in the image domain, few-shot video classification is relatively unexplored. We argue that previous methods underestimate the importance of video feature learning and propose to learn spatiotemporal features using a 3D CNN. Proposing a two-stage approach that learns video features on base classes followed by fine-tuning the classifiers on novel classes, we show that this simple baseline approach outperforms prior few-shot video classification methods by over 20 points on existing benchmarks. To circumvent the need of labeled examples, we present two novel approaches that yield further improvement. First, we leverage tag-labeled videos from a large dataset using tag retrieval followed by selecting the best clips with visual similarities. Second, we learn generative adversarial networks that generate video features of novel classes from their semantic embeddings. Moreover, we find existing benchmarks are limited because they only focus on 5 novel classes in each testing episode and introduce more realistic benchmarks by involving more novel classes, i.e. few-shot learning, as well as a mixture of novel and base classes, i.e. generalized few-shot learning. The experimental results show that our retrieval and feature generation approach significantly outperform the baseline approach on the new benchmarks.